Latency-constrained DNN architecture learning for edge systems using zerorized batch normalization

Latency-constrained DNN architecture learning for edge systems using zerorized batch normalization

Deep learning applications have been widely adopted on edge devices, to mitigate the privacy and latency issues of accessing cloud servers. Deciding the number of neurons during the design of a deep neural network to maximize performance is not intuitive. Particularly, many application scenarios are...

Fuld beskrivelse

Bibliografiske detaljer
Main Authors: Huai, Shuo, Liu, Di, Kong, Hao, Liu, Weichen, Subramaniam, Ravi, Makaya, Christian, Lin, Qian
Andre forfattere: School of Computer Science and Engineering
Format: Journal Article
Sprog:English
Udgivet: 2023
Fag:
Engineering::Computer science and engineering::Software::Software engineering
Latency Optimization
Edge Device
Latency Prediction
Online adgang:https://hdl.handle.net/10356/165565

Internet

https://hdl.handle.net/10356/165565

Lignende værker